llvm
/
torch-mlir
mirror of https://github.com/llvm/torch-mlir
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Remove last mentions of IREE.

pull/346/head
Sean Silva 2021-10-01 17:27:42 +00:00
parent 9fc059e948
commit 5917f1dc47
2 changed files with 4 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
include/torch-mlir/Conversion/Passes.td
View File

@ -93,7 +93,7 @@ def ConvertTorchToLinalg : Pass<"convert-torch-to-linalg", "FuncOp"> {
4. All this code operates on ranked tensors, for which using individual 4. All this code operates on ranked tensors, for which using individual
SSA values for sizes (rather than a "shape type") seems to SSA values for sizes (rather than a "shape type") seems to
work really well at this level of abstraction based on prior experience work really well at this level of abstraction based on prior experience
in IREE. (unranked code tends to benefit from having a discrete in other projects. (unranked code tends to benefit from having a discrete
"shape type" to model shapes). "shape type" to model shapes).
We will see if we end up needing something like `shape.assuming`, but for We will see if we end up needing something like `shape.assuming`, but for

8
python/torch_mlir_e2e_test/linalg_on_tensors_backends/refbackend.py
View File

@ -88,7 +88,7 @@ LOWERING_PIPELINE = ",".join([
class RefBackendLinalgOnTensorsBackend(LinalgOnTensorsBackend): class RefBackendLinalgOnTensorsBackend(LinalgOnTensorsBackend):
"""Main entry-point for the backend.""" """Main entry-point for the reference backend."""
def __init__(self): def __init__(self):
super().__init__() super().__init__()
@ -102,10 +102,8 @@ class RefBackendLinalgOnTensorsBackend(LinalgOnTensorsBackend):
imported_module: The MLIR module consisting of funcs in the torch imported_module: The MLIR module consisting of funcs in the torch
dialect. dialect.
Returns: Returns:
An opaque, backend specific module object that can be passed to load. An opaque, backend specific compiled artifact object that can be
The object may actually be something more specific to the backend (i.e. passed to `load`.
for IREE, it is a serialized VM flatbuffer) but the contract is that
it is operated on by methods on this class.
""" """
with imported_module.context: with imported_module.context:
pm = PassManager.parse(LOWERING_PIPELINE) pm = PassManager.parse(LOWERING_PIPELINE)